An error I have never seen before

[Jammy]

Has anybody ever seen this error? I haven't seen it in the past, and it is preventing me from finishing this test on my XP 2600:

http://members.arstechnica.com/x/jamilla/errorSB.jpg

I will await an answer.

Thanks,

Jammy


http://www.clanhosts.com/dev/sobsig/sig.php?Jammy

[Jammy]

My client on the XP 3000 is crashing now!


http://members.arstechnica.com/x/jamilla/program_errorXP3000.gif



It seems that all that is running noiw are the two laptops, mine and my daughters'. Both have XP as OS. The others are W2K.

Jammy

[Jammy]

And I still have my logs from both boxes.

Jammy

[vjs]

Jammy,

Could this mean the first prime for TPR??? (Divide by Zero) :p

[Jammy]

Originally posted by vjs
Jammy,

Could this mean the first prime for TPR??? (Divide by Zero) :p

How do you figure?

[MikeH]

If you'd like to give an older version a try to see if it's specifically a problem 1.2.5, you can download

1.2.0. (http://www.aooq73.dsl.pipex.com/tmp/sb120.exe)
1.1.0. (http://www.aooq73.dsl.pipex.com/tmp/sb110.exe)
1.0.0. (http://www.aooq73.dsl.pipex.com/tmp/sb100.exe)

I'd suggest you take a copy of the existing SB folder and note down the current k and n, but installing one of these over the top of the existing version should be OK, and it should just continue processing the existing unit.

Please report back.

If nothing works you could always try 'secret' or 'garbage' for a short while until things are fixed - these are much smaller numbers n ~ 1M and ~ 4M respectively (but won't score for you or your team), here (http://www.seventeenorbust.com/secret/) for more info.

[Jammy]

Originally posted by MikeH
If you'd like to give an older version a try to see if it's specifically a problem 1.2.5, you can download

I have all of the older versions, thanks.



I'd suggest you take a copy of the existing SB folder and note down the current k and n, but installing one of these over the top of the existing version should be OK, and it should just continue processing the existing unit.

I started a new test on this box (XP 2600) and the XP 3000. I saved the logs and the z-file. I do not sieve so if anybody would like them they are most welcome to them!


Please report back.

If nothing works you could always try 'secret' or 'garbage' for a short while until things are fixed - these are much smaller numbers n ~ 1M and ~ 4M respectively (but won't score for you or your team), here (http://www.seventeenorbust.com/secret/) for more info.

Sorry I am just a cruncher and do not undertstand any of those "Server Assignment Queues".

As I said above, anybody is welcome to try to finish those tests or use them in any way that will benefit Seventeen or Bust!

Jammy

[jjjjL]

Very interesting. Send me the z-files. lhelm@seventeenorbust.com

I'll fix this for you. Thanks Jammy.

Cheers,
Louie

[Jammy]

Originally posted by jjjjL
Very interesting. Send me the z-files. lhelm@seventeenorbust.com

I'll fix this for you. Thanks Jammy.

Cheers,
Louie

Thanks Louie! They are on their way.


Jammy

[jjjjL]

I found the offending code that can sometimes divide by zero.

It has to do with the ETA calculation not any of the important math. I can make a fix but I'm going to investigate a litte first to see how it is this situation can come about in the first place.

This seems to only happen on a restart when zero blocks are done and the save file is in a goofy state.

A workaround for now is to turn off transmission of intermediate blocks transmission for 1 block. You should be able to turn it back on after that.

Also, I will finish these two tests on my 3.4GHz P4 under your account Jammy. The first is done already. Not a prime... maybe the second. ;) It's still your prime if it is.

Cheers,
Louie

[Jammy]

Originally posted by jjjjL
I found the offending code that can sometimes divide by zero.

It has to do with the ETA calculation not any of the important math. I can make a fix but I'm going to investigate a litte first to see how it is this situation can come about in the first place.

This seems to only happen on a restart when zero blocks are done and the save file is in a goofy state.

A workaround for now is to turn off transmission of intermediate blocks transmission for 1 block. You should be able to turn it back on after that.

Also, I will finish these two tests on my 3.4GHz P4 under your account Jammy. The first is done already. Not a prime... maybe the second. ;) It's still your prime if it is.

Cheers,
Louie

Noted! If I ever have to turn off a computer I will stop intermediate block transmissions for awhile when I turn it back on.

And thanks again Louie! I saw how fast those two tests were finished.

Jammy

[Joh14vers6]

Originally posted by jjjjL
I found the offending code that can sometimes divide by zero.

It has to do with the ETA calculation not any of the important math. I can make a fix but I'm going to investigate a litte first to see how it is this situation can come about in the first place.

This seems to only happen on a restart when zero blocks are done and the save file is in a goofy state.

A workaround for now is to turn off transmission of intermediate blocks transmission for 1 block. You should be able to turn it back on after that.

Cheers,
Louie

Louie, what is progress state of fixing this problem.

Because, we (DPC-members) also getting this problem with v2.0 SSE2.
Only when the client is running with the servicehandler we get this errorreport:

0000: 41 70 70 6c 69 63 61 74 Applicat
0008: 69 6f 6e 20 46 61 69 6c ion Fail
0010: 75 72 65 20 20 73 62 2e ure sb.
0018: 65 78 65 20 30 2e 30 2e exe 0.0.
0020: 30 2e 30 20 69 6e 20 73 0.0 in s
0028: 62 2e 65 78 65 20 30 2e b.exe 0.
0030: 30 2e 30 2e 30 20 61 74 0.0.0 at
0038: 20 6f 66 66 73 65 74 20 offset
0040: 30 30 30 31 39 30 31 32 00019012

See also my first posting (http://www.free-dc.org/forum/showthread.php?threadid=7618) about this problem.

http://home.hccnet.nl/ajw.van.rhenen/Error.png

[Joh14vers6]

I got this error in the mid of a test:

0000: 41 70 70 6c 69 63 61 74 Applicat
0008: 69 6f 6e 20 46 61 69 6c ion Fail
0010: 75 72 65 20 20 53 62 2e ure Sb.
0018: 65 78 65 20 30 2e 30 2e exe 0.0.
0020: 30 2e 30 20 69 6e 20 53 0.0 in S
0028: 62 2e 65 78 65 20 30 2e b.exe 0.
0030: 30 2e 30 2e 30 20 61 74 0.0.0 at
0038: 20 6f 66 66 73 65 74 20 offset
0040: 30 30 32 32 35 62 39 35 00225b95

[Joh14vers6]

Originally posted by jjjjL
I found the offending code that can sometimes divide by zero.

It has to do with the ETA calculation not any of the important math. I can make a fix but I'm going to investigate a litte first to see how it is this situation can come about in the first place.

This seems to only happen on a restart when zero blocks are done and the save file is in a goofy state.

A workaround for now is to turn off transmission of intermediate blocks transmission for 1 block. You should be able to turn it back on after that.

Also, I will finish these two tests on my 3.4GHz P4 under your account Jammy. The first is done already. Not a prime... maybe the second. ;) It's still your prime if it is.

Cheers,
Louie
I still have problems with this. I tried the workaround, but have troubles to get back on transmitting.

All clients are running with sobsvc with interact with desktop off. Standard I have set the registry at [HKEY_LOCAL_MACHINE\SOFTWARE\LhDn\sob\config]
"transmit"=dword:00000000 after some time I change the registry to [HKEY_LOCAL_MACHINE\SOFTWARE\LhDn\sob\config]
"transmit"=dword:00000001 and for shutdown of the machine I set it back to [HKEY_LOCAL_MACHINE\SOFTWARE\LhDn\sob\config]
"transmit"=dword:00000000.
I can see in the logs that when I switch transmitting back on the client is not transmitting at all.

Within the GUI it is working, but I have no GUI of course and I am not sitting behind those machines.